Censoring with Plausible Deniability: Asymmetric Local Privacy for Multi-Category CDF Estimation

Qirui Hu, Yi Liu
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:45290-45311, 2026.

Abstract

We introduce a new mechanism within the Utility-Optimized Local Differential Privacy (ULDP) framework that enables censoring with plausible deniability when collecting and analyzing sensitive data. Our approach addresses scenarios where certain values, such as large numerical responses, are more privacy-sensitive than others, while accompanying categorical information may not be private on its own but could still be identifying. The mechanism selectively withholds identifying details when a response might indicate sensitive content, offering asymmetric privacy protection. Unlike previous methods, it avoids the need to predefine an exact sensitive region, making it more adaptable and practical. Although the mechanism is designed for ULDP, it can also be applied under symmetric LDP settings, where it still benefits from censoring and reduced perturbation cost. We provide theoretical guarantees, including uniform consistency and pointwise weak convergence results. Numerical experiments on both synthetic data and real-world data demonstrate the validity of the proposed methodology.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-hu26ae, title = {Censoring with Plausible Deniability: Asymmetric Local Privacy for Multi-Category {CDF} Estimation}, author = {Hu, Qirui and Liu, Yi}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {45290--45311}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/hu26ae/hu26ae.pdf}, url = {https://proceedings.mlr.press/v306/hu26ae.html}, abstract = {We introduce a new mechanism within the Utility-Optimized Local Differential Privacy (ULDP) framework that enables censoring with plausible deniability when collecting and analyzing sensitive data. Our approach addresses scenarios where certain values, such as large numerical responses, are more privacy-sensitive than others, while accompanying categorical information may not be private on its own but could still be identifying. The mechanism selectively withholds identifying details when a response might indicate sensitive content, offering asymmetric privacy protection. Unlike previous methods, it avoids the need to predefine an exact sensitive region, making it more adaptable and practical. Although the mechanism is designed for ULDP, it can also be applied under symmetric LDP settings, where it still benefits from censoring and reduced perturbation cost. We provide theoretical guarantees, including uniform consistency and pointwise weak convergence results. Numerical experiments on both synthetic data and real-world data demonstrate the validity of the proposed methodology.} }
Endnote
%0 Conference Paper %T Censoring with Plausible Deniability: Asymmetric Local Privacy for Multi-Category CDF Estimation %A Qirui Hu %A Yi Liu %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-hu26ae %I PMLR %P 45290--45311 %U https://proceedings.mlr.press/v306/hu26ae.html %V 306 %X We introduce a new mechanism within the Utility-Optimized Local Differential Privacy (ULDP) framework that enables censoring with plausible deniability when collecting and analyzing sensitive data. Our approach addresses scenarios where certain values, such as large numerical responses, are more privacy-sensitive than others, while accompanying categorical information may not be private on its own but could still be identifying. The mechanism selectively withholds identifying details when a response might indicate sensitive content, offering asymmetric privacy protection. Unlike previous methods, it avoids the need to predefine an exact sensitive region, making it more adaptable and practical. Although the mechanism is designed for ULDP, it can also be applied under symmetric LDP settings, where it still benefits from censoring and reduced perturbation cost. We provide theoretical guarantees, including uniform consistency and pointwise weak convergence results. Numerical experiments on both synthetic data and real-world data demonstrate the validity of the proposed methodology.
APA
Hu, Q. & Liu, Y.. (2026). Censoring with Plausible Deniability: Asymmetric Local Privacy for Multi-Category CDF Estimation. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:45290-45311 Available from https://proceedings.mlr.press/v306/hu26ae.html.

Related Material